Immunohistochemical expression of MAM-3 and MAM-6 antigens in salivary gland tumours

Summary MAM-3 and MAM-6 antigens of human milk fat globule membrane were detected immunohistochemically in 93 cases of salivary gland tumours as well as in normal glands. The antigens were visualized in 10% formalin-fixed paraffin sections. MAM-3 (MoAbs 115G3, 67D11) antigen was distributed in intercalated and striated duct cells of the normal salivary glands, and in luminal tumour cells and squamous metaplastic cells of pleomorphic adenomas. In pleomorphic adenomas the frequency of positive staining with MoAb 67D11 (54/67; 80.6%) was higher than that with MoAb 115G3 (36/67; 53.7%). MAM-6 (MoAbs 115D8, 115F5) antigen was expressed in luminal and lateral borders of serous acinar cells and ductal of the normal glands, and also in luminal borders of tubulo-ductal and glandular structures of salivary gland tumours. Ductal basal cells were characterized by existence of positive staining for MAM-6 antigen, in adenolymphomas MAM-6 antigen was restricted to the basal tumour cells. Some mucous cells of mucoepidermoid tumours were stained specifically with MoAb 115G3, and epidermoid cells of mucoepidermoid carcinomas manifested MAM-6 antigen staining. Immunohistochemical localization of MAM-6 antigen resembled that of epithelial membrane antigen (EMA) detected with MoAb.     

Keratin and vimentin distribution patterns in the epithelial structures of the cannie anal region

The intermediate filament labeling pattern of the epithelial structures of the canine anal region was studied with different polypeptide specific keratin monoclonal antibodies (MoAbs) and with a monoclonal and polyclonal vimentin antibody. The epithelial structures in this region could be discriminated and characterized by differences in their keratin staining pattern. The basal cells in the different epithelial structures showed a similar staining pattern characterized by reactivity with MoAbs staining keratins 5, 8, 14, and 17. Columnar epithelial cells showed a completely different phenotype mostly characterized by reactivity with MoAbs staining keratins 7, 5, 8, 18, and 19. A restricted number of differentiated perianal gland cells showed perinuclear vimentin staining. Myoepithelial cells did not stain for vimentin, but, as other basal cells, were positive for MoAbs staining keratins 5, 8, 14, and 17.© Willey-Liss, Inc.     

Keratin and vimentin distribution patterns in the epithelial structures of the canine anal region.

The intermediate filament labeling pattern of the epithelial structures of the canine anal region was studied with different polypeptide specific keratin monoclonal antibodies (MoAbs) and with a monoclonal and polyclonal vimentin antibody. The epithelial structures in this region could be discriminated and characterized by differences in their keratin staining pattern. The basal cells in the different epithelial structures showed a similar staining pattern characterized by reactivity with MoAbs staining keratins 5, 8, 14, and 17. Columnar epithelial cells showed a completely different phenotype mostly characterized by reactivity with MoAbs staining keratins 7, 5, 8, 18, and 19. A restricted number of differentiated perianal gland cells showed perinuclear vimentin staining. Myoepithelial cells did not stain for vimentin, but, as other basal cells, were positive for MoAbs staining keratins 5, 8, 14, and 17.     

Expression of alpha-methylacyl-CoA racemase (P504s) in various malignant neoplasms and normal tissues: astudy of 761 cases

Alpha-methylacyl CoA racemase (AMACR), also known as P504S, plays an important role in peroxisomal beta-oxidation of branched-chain fatty acids. It has recently been shown that AMACR is highly expressed in prostate cancer and that it may be an important diagnostic marker for prostate carcinoma. However, little is known about expression of AMACR in normal tissues and other malignant tumors. In this study, we investigated expression of AMACR in 539 malignant tumors and 222 normal human tissues of various types by immunohistochemical analysis. mRNA levels of AMACR in normal organs and in selected tumors were assessed by real time PCR. In normal tissue, high expression of AMACR mRNA was identified in liver, kidney and salivary gland, while AMACR protein was detected in liver (hepatocytes), kidney (tubular epithelial cells), lung (only bronchial epithelial cells), and gallbladder (only mucosal epithelial cells). High expression of AMACR mRNA was found in prostate, liver, and kidney cancers but rarely in stomach and bladder cancers. A high percent of adenocarcinomas arising from these organs express AMACR, including 17 of 21 (81%) of hepatocellular carcinomas and 18 of 24 (75%) of renal cell carcinomas. In addition, carcinomas arising from tissues normally not expressing AMACR were also positive for the antigen, including 17 of 18 (94%) prostate carcinomas, 9 of 29 (31%) of urothelial carcinomas, and 4 of 15 (27%) of gastric adenocarcinomas. Two hundred and fifty cases of adenocarcinomas from lung, breast, pancreas, bile duct, adrenal gland, salivary gland, ovary, thyroid and endometrium were negative or rarely positive for AMACR. Neuroendocrine carcinomas rarely expressed AMACR. Melanomas, squamous cell carcinomas, basal cell carcinomas, soft tissue tumors (including epithelioid sarcomas and synovial sarcoma), thymomas, and germ cell tumors were negative for AMACR. Our data provide important baseline information for using AMACR in clinical practice and also are valuable in furthering understanding of the pathogenic role of AMACR in malignant neoplasms.     

Immunohistochemical analysis of the distribution of a breast tumor associated antigen recognized by a monoclonal antibody.

A new monoclonal antibody (MoAb), MM 1-80, recognizing a tumor associated epitope of a breast high molecular weight mucin molecule was tested, using the avidin biotin immunoperoxidase method on normal and pathological mammary tissues. The normal mammary ducts and lobules were negative. Fibroadenomas showed a strong intracytoplasmic staining. In apocrine metaplasia, adenosis, and papillomatosis, scattered cells showed intracytoplasmic, luminal border or secretion reactivity. In lobular and ductal hyperplasia the cells showed intracytoplasmic immunoreactivity which, however, became more intense and homogeneous in atypical lesions, i.e. lobular and ductal in-situ carcinomas. The infiltrating carcinomas of different histotype expressed positivity on 98% of the cases (113/115) and axillary metastatic lymph nodes were always positive (20/20). The MoAb was tested on 175 human neoplasias of different origin which were in the majority of the cases negative with the exception of adenocarcinomas of the lung, ovary and bladder. MM1-80 appears to react preferentially with mammary cells undergoing hyperplastic, metaplastic and neoplastic processes. The 1-80 epitope distribution is different in these lesions starting with a predominant luminal expression in benign lesions and becoming strong and cytoplasmic in the malignant breast cell.     

Identification of a Vimentin-Like Function Associated Molecule (FAM) on Rat NK Cells: Evidence for Receptor Function

Monoclonal antibody (MoAb) 5C6 specifically binds to fish, rat and human NK cells and inhibits cytotoxicity. The molecule recognized by this MoAb is a 50–53–kDa membrane protein on rat leukaemic NK (CRC) cells. In the present study, we have obtained a partial internal amino acid sequence from a purified 42-kDa fragment of the CRC-function associated molecule (FAM). Three tryptic peptide fragments were sequenccd and each showed homology to intermediate filament vimentin sequences as deduced from (GenBank) mouse cDNA sequences. Amino acid composition analysis indicated that similar to cytoskeletal vimentin, the FAM contained a high percentage of non-polar amino acids. To further assess the similarities between this protein and vimentin., two commercially available anti-vimentin MoAbsand one anti-vimentin polyclonal antibody were tested for binding and inhibition of NK cytotoxicity. All anti-vimentin MoAbs inhibited killing by rat NWNA cells of appropriate targets. Anti-vimentin MoAb 13.2 bound to 41% of NWNA cells compared with approximately 58% binding for MoAb 5C6. Capping and sequential binding experiments showed that MoAb 5C6 effectively removed, from CRC-cell membranes, the protein recognized by MoAb V9. Sequential addition of these two MoAbs (MoAb 13.2 followed by MoAb V9) to CRC cells did not produce competitive binding. Biochemical and Western blot analysis of the vimentin-like protein obtained from CRC cells indicated that this protein has a molecular weight of 48–50 kDa, with an isoelectric point of pH 6.1–6.3. This protein is cross-reactive by Western blot analysis with anti-vimentin and anti-intermediate filament (IFA) antigen MoAbs but not with anti-desmin or anti-actin M oAbs. The molecular weight heterogeneity (43 versus 48 50 k Da) of the CRC protein was also examined. Western blot analysis of the CRC extract after different in vitro incubation times at 37°C and 4°C demonstrated that the 50 53–kDa ‘native’ protein degraded to a 42-kDa protein by 24 and 48 h respectively. This degradation was inhibitable by 10 mm EGTA. Evidence is presented which indicates that a vimentin-like protein on transformed rat NK cells may be an antigen binding receptor which initiates target cell lysis.     

Expression of antigen reactive with a monoclonal antibody to HTLV-1 P19 in salivary glands in Sjögren''s syndrome.

To examine the possible involvement of retroviruses in Sjögren's syndrome (SS), labial salivary gland sections from 99 individuals were probed with three MoAbs to core (gag) proteins of human T cell leukaemia virus-1 (HTLV-1) and two MoAbs to HIV-1. Sections from 31% of 39 patients with primary SS (pSS) contained an epithelial cytoplasmic protein reactive with a MoAb(197) to the p19 group specific antigen (gag) of HTLV-1. The antigen was also detected in samples from 24% of 17 patients with rheumatoid arthritis (RA) and SS. 21% of 14 patients with sicca symptoms and 12.5% of 16 patients with other connective tissue diseases. It was not found in the salivary glands of 13 normal controls. A second MoAb to p19 gag, a MoAb to the p24 gag of HTLV-1 and MoAbs to HIV-1 p17and p24 gags gave negative reactions. Serum antibodies to HTLV-1 were negative, confirming that the antigen was not part of HTLV-1. The antigen showed properties consistent with an endogenous retrovirus in that it was absent in healthy tissues or resting cells but inducible by stimulation with phytohaemagglutinin (PHA) or interferon-gamma (IFN-γ) It appeared to be distinct from the endogenous retroviral sequence HRES-1. These data suggest the presence of an endogenous retrovirus in salivary gland epithelium which could contribute to the chronic inflammation of SS.     

Immunohistochemical and biochemical characterization of the mucin-type tumor associated antigen TAG-12 by monoclonal antibody 7A9.

The reactivity of monoclonal antibody 7A9 with normal and neoplastic human tissues and some biochemical characteristics of the antigen (TAG-12) bound by 7A9 were evaluated. Antibody 7A9 showed broad reactivity with various carcinomas in paraffin sections. High percentages of positive tumor cells, displaying membrane and cytoplasmic staining, were noticed in adenocarcinomas of the breast (83/85), serous cystadenocarcinomas of the ovary (15/16), and lung adenocarcinomas (14/16). TAG-12 antigen was also detectable in normal adult and fetal tissues, but the reactivity of 7A9 was mainly restricted to the luminal surface of epithelial cells. For biochemical analyses, TAG-12 antigen purified from T47-D breast carcinoma cells by lectin affinity chromatography was treated with different glycosidases and proteases and analyzed by immunoblotting with 7A9. The data indicate that the antigen recognized by 7A9 is a heavily sialated mucin-type glycoprotein with a molecular weight of more than 200 KD. Similar to all other antibodies against tumor associated antigens, monoclonal antibody 7A9 is not tumor-specific but displays tissue staining patterns with a high carcinoma-to-normal ratio. The strong reactivity with the majority of tumor cells in several carcinoma types suggests that 7A9 is useful for in vitro and in vivo targeting of those tumors.     

Gastric phenotypic expression in human gallbladder cancers revealed by pepsinogen immunohistochemistry and mucin histochemistry

Summary Gastric phenotypic expression indicated by paradoxical concanavalin A (Con A) staining for class III mucins and the immunoperoxidase method for pepsinogen (Pg) I and Pg II was found in pyloric gland metaplasia of gallbladder epithelium. Using the same methods, the features of gallbladder cancers and their relationship to pyloric gland metaplasia in the human gallbladder epithelium were studied. Histologically, 57 gallbladder cancers were classified into 5 papillary adenocarcinomas, 29 tubular adenocarcinomas, 8 poorly differentiated adenocarcinomas, 6 signet-ring cell carcinomas, 4 mucinous adenocarcinomas, and 5 squamous cell carcinomas. In papillary and tubular adenocarcinomas, Pg I and/or Pg II staining was detected in 80% and 75.9% of cancers, respectively. Pg II staining was significantly more frequent than Pg I staining. One signetring cell carcinoma also had Pg II activity. Pyloric gland metaplasias all contained class III mucins and were further classified into complete type and incomplete type on the basis of presence or absence Pg I and/ or Pg II activities. A few cancer cells with class III mucins were negative for Pg staining; conversely, a few cells with Pg I and/or Pg II had no class III mucins. Phenotypic diversity in both class III mucin reactivity and Pg activities was observed in gallbladder cancer cells with the pyloric gland cell type. By comparison, pyloric gland metaplasia varied only in Pg activities. A few Pg-positive cancers were found in the gallbladder with Pg-negative pyloric gland metaplasia. The present results clearly indicate the appearance of gastric phenotypic expression in both gallbladder epithelium and gallbladder cancers and suggest the independent induction of pyloric gland metaplasia and cancer with gastric phenotypic expression.     

Distribution of vitamin B12 R-binder in lung tumors. Implications for cell differentiation

Expression of vitamin B12 R-binder, a specific binding protein for vitamin B12, was studied immunohistochemically in normal lung tissues and 107 lung tumors of various types. In normal tissues, vitamin B12 R-binder (R-binder) expression was restricted to the mucous cells of bronchial or bronchiolar epithelium and submucosal glands as well as to nonciliated bronchiolar (Clara) cells. Among lung carcinomas, 38% of squamous cell carcinomas, 42% of adenocarcinomas and 23% of large cell carcinomas showed positive staining for R-binder whereas small cell carcinomas did not. These findings offer the possibility that a majority of the histologic types of lung carcinoma have common histogenetical characteristics with mucous or Clara cells. Of the bronchial gland tumors, R-binder could be detected in a mucoepidermoid carcinoma but not in adenoid cystic carcinomas. Epithelial components in both pulmonary blastomas and hamartomas showed a reactivity for R-binder, suggesting that these tumors contained components composed of cells with bronchiolar cell differentiation. The immunohistochemical examination of lung tumors, using anti-R-binder antibody, may have some implications in the cell differentiation of lung tumors.     

Immunoreactivity with monoclonal antibody A-80 and nuclear DNA content in benign and malignant human breast disease.

Immunohistochemical analysis with the monoclonal antibody (MoAb) A-80, recognizing a tumor-associated cytoplasmic mucin-type glycoprotein, and cytometric nuclear DNA assessment were performed on 314 surgical specimens of the human mammary gland. The series included 36 benign conditions, 34 epithelial hyperplasias, 40 carcinomas in situ, and 204 primary invasive carcinomas. Normal breast parenchyma, benign tumors, and other nonmalignant lesions were all of DNA diploid (euploid) type and rarely expressed the A-80 glycoprotein. Differences in MoAb A-80 immunoreactivity and nuclear DNA content were noted among subtypes of epithelial hyperplasias. Fifteen of 34 epithelial hyperplasias were of DNA aneuploid type and the majority were A-80 immunoreactive. Of these 15 immunoreactive aneuploid epithelial hyperplasias, atypical intraductal hyperplasia was the most common subgroup. None of the 19 epithelial hyperplasias of DNA euploid type immunoreacted. Most of the intraductal (33 of 40) and invasive (180 of 204) carcinomas immunostained with MoAb A-80. The majority of the A-80 immunoreactive malignant tumors were of DNA aneuploid type (26 of 33 carcinomas in situ and 108 of 180 invasive mammary carcinomas). The results suggest that expression of the A-80 glycoprotein occurs at an early stage of malignant transformation. Genetically stable (euploid) mammary tumors seem to immunoreact with MoAb A-80 less frequently than genetically unstable (aneuploid) tumor variants. Combined analysis with MoAb A-80 and of nuclear DNA content in premalignant and malignant mammary lesions could be a useful tool of differential diagnostic and prognostic value.     

Expression and activity of transglutaminase II in spontaneous tumours of dogs and cats

Tissue transglutaminase II (TGase II) is a dual function protein with both transamidating and guanidine triphosphate (GTP)-binding capabilities. Previous studies have implicated TGase as a pro-apoptotic molecule; however, our recent findings indicate that TGase II may act as a survival factor in various cell types. The purpose of this study was to survey TGase II expression in normal tissue and spontaneous tumours of dogs and cats, by Western blotting and immunohistochemistry. Bladder, liver and adrenal gland exhibited prominent expression of TGase II while other tissues, including mammary gland, displayed limited expression and activity. TGase II GTP-binding in normal tissues was proportional to the level of expression in all tissues examined. Normal mammary tissue and that showing benign hyperplasia did not express TGase II. However, 11/25 (44%) of canine mammary carcinomas and 10/12 (83%) of feline mammary carcinomas strongly expressed TGase II in either a stromal, cellular or combined pattern. The pattern of expression was not related to the classification of mammary carcinoma (solid, tubulopapillary, complex or anaplastic), except that two anaplastic canine mammary carcinomas showed prominent TGase II expression. Two canine mammary carcinoma cell lines showed prominent TGase expression, and when the TGase activity was inhibited, the cells became more sensitive to doxorubicin-induced cell death. Thus, TGase II was significantly expressed in mammary cancers from dogs and cats and immunoreactivity of TGase II was similar to that reported in humans beings. The pro-survival effect of TGase II in canine mammary carcinoma cell lines was similar to that previously reported in humans patients.     

Immunohistochemical demonstration of nonspecific cross-reacting antigen in normal and neoplastic human tissues using a monoclonal antibody. Comparison with carcinoembryonic antigen localization

Nonspecific cross-reacting antigen (NCA) immunoreactivity was localized in normal and neoplastic human tissues using a monoclonal antibody to 55, 90 and 95 kDa molecules of NCA. This was compared to the localization of immunoreactive carcinoembryonic antigen (CEA) as demonstrated by polyclonal and monoclonal antibodies. In frozen sections, CEA was localized in normal surface epithelium of the stomach and colon where NCA was only weakly detected. Type 1 and type 2-like pneumocytes were positive for NCA, while CEA was localized only in type 2-like pneumocytes. CEA and NCA were both demonstrated in ductal cells of frozen pancreatobiliary and mammary tissues. The antigenicity of CEA and NCA in normal tissues was significantly lost after paraffin embedding as compared to frozen sections. NCA was consistently demonstrated in eccrine sweat glands embedded in paraffin. In various tumor tissues, CEA and NCA were colocalized and expression increased sufficiently to be detected in paraffin sections. Adenocarcinomas of the stomach and colon and cystadenocarcinoma of the pancreas, as well as neuroendocrine carcinomas of the lung and thyroid, showed a CEA predominance over NCA. In ductal adenocarcinomas of the pancreas and breast and in cholangiocarcinoma, NCA reactivity was greater than CEA. Keratinizing foci of most squamous cell carcinomas of mucosal origin and some adenocarcinomas equally expressed both. Hepatocellular carcinoma, lobular mammary carcinoma and papillary thyroid carcinoma were positive only with unabsorbed polyclonal antibody which widely recognizes CEA-related substances. Renal cell carcinoma, prostatic adenocarcinoma, transitional cell carcinoma, anaplastic carcinomas, choriocarcinoma and basal cell carcinomas showed little or no immunoreactivity. Hence the relative ratio of CEA/NCA expression in tumors was dependent on the tissue of origin and histologic type. The cytoplasmic granular staining of NCA in cancer cells was a noteworthy difference from the plasma membrane-associated localization of CEA.     

Immunohistochemical Demonstration of Nonspecific Cross-reacting Antigen in Normal and Neoplastic Human Tissues Using a Monoclonal Antibody: Cornparison with Carcinoembryonic Antigen Localization

Nonspecific cross-reacting antigen (NCA) immunoreactivity was localized in normal and neoplastic human tissues using a monoclonal antibody to 55, 90 and 95 kDa molecules of NCA. This was compared to the localization of immunoreactive carcinoembryonic antigen (CEA) as demonstrated by polyclonal and monoclonal antibodies. In frozen sections, CEA was localized in normal surface epithelium of the stomach and colon where NCA was only weakly detected. Type 1 and type 2 like pneumocytes were positive for NCA, while CEA was localized only in type 2-like pneumocytes. CEA and NCA were both demonstrated in ductal cells of frozen pancreatobiliary and mammary tissues. The antigenicity of CEA and NCA in normal tissues was significantly lost after paraffin embedding as compared to frozen sections. NCA was consistently demonstrated in eccrine sweat glands embedded in paraffin. In various tumor tissues, CEA and NCA were colocalized and expression increased sufficiently to be detected in paraffin sections. Adenocarcinomas of the stomach and colon and cystadenocarcinoma of the pancreas, as well as neuroendocrine carcinomas of the lung and thyroid, showed a CEA predominance over NCA. In ductal adenocarcinomas of the pancreas and breast and in cholangiocarcinoma, NCA reactivity was greater than CEA. Keratiniring foci of most squamous cell carcinomas of mucosal origin and some adenocarcinomas equally expressed both. Hepatocellular carcinoma, lobular mammary carcinoma and papillary thyroid carcinoma were positive only with unabsorbed polyclonal antibody which widely recognizes CEA-related substances. Renal cell carcinoma, prostatic adenocarcinoma, transitional cell carcinoma, anaplastic carcinomas, choriocarcinoma and basal cell carcinomas showed little or no immunoreactivity. Hence the relative ratio of CEA/NCA expression in tumors was dependent on the tissue of origin and histologic type. The cytoplasmic granular staining of NCA in cancer cells was a noteworthy difference from the plasma membrane-associated localization of CEA. Acta Pathol Jpn 40: 85–97, 1990.     

p40 (ΔNp63) is superior to p63 for the diagnosis of pulmonary squamous cell carcinoma

Immunohistochemistry has recently emerged as a powerful ancillary tool for differentiating lung adenocarcinoma and squamous cell carcinoma-a distinction with important therapeutic implications. Although the most frequently recommended squamous marker p63 is extremely sensitive, it suffers from low specificity due to its reactivity in a substantial proportion of lung adenocarcinomas and other tumor types, particularly lymphomas. p40 is a relatively unknown antibody that recognizes ΔNp63-a p63 isoform suggested to be highly specific for squamous/basal cells. Here we compared the standard p63 antibody (4A4) and p40 in a series of 470 tumors from the archives of Memorial Sloan-Kettering Cancer Center and The Johns Hopkins Hospital, which included lung squamous cell carcinomas (n=81), adenocarcinomas (n=237), and large cell lymphomas (n=152). The p63 was positive in 100% of squamous cell carcinomas, 31% of adenocarcinomas, and 54% of large cell lymphomas (sensitivity 100%, specificity 60%). In contrast, although p40 was also positive in 100% of squamous cell carcinomas, only 3% of adenocarcinomas, and none of large cell lymphomas had p40 labeling (sensitivity 100%, specificity 98%). The mean percentage of p63 versus p40-immunoreactive cells in squamous cell carcinomas was equivalent (97 vs 96%, respectively, P=0.73). Rare adenocarcinomas with p40 labeling had reactivity in no more than 5% of tumor cells, whereas the mean (range) of p63-positive cells in adenocarcinomas and lymphomas was 26% (1-90%) and 48% (2-100%), respectively. In summary, p40 is equivalent to p63 in sensitivity for squamous cell carcinoma, but it is markedly superior to p63 in specificity, which eliminates a potential pitfall of misinterpreting a p63-positive adenocarcinoma or unsuspected lymphoma as squamous cell carcinoma. These findings strongly support the routine use of p40 in place of p63 for the diagnosis of pulmonary squamous cell carcinoma.     

Reactivity of a monoclonal antibody with tissues and tumors from the human breast. Immunohistochemical localization of a new antigen and clinicopathologic correlations.

The reactions of a monoclonal antibody to the MCF7 breast cancer cell line were immunohistochemically studied on a variety of breast tumors, primary and metastatic, on mammary epithelium and on nonneoplastic breast lesions. A high proportion of positive reactions was observed in ductal, lobular, and tubular carcinomas as well as in mammary Paget's disease. Mucinous, medullary, and papillary carcinomas showed a low incidence of reactivity. Carcinomas with metaplasia, carcinoids, and nonepithelial breast tumors were unreactive with the antibody. Positive immunostaining was documented also in nodal and extranodal metastatic lesions. The staining of nodal metastases was correlated with the positive reaction of the primary tumor. Reactivity was widely distributed in normal breast epithelial cells and in benign breast lesions. Staining of nonneoplastic mammary epithelial was associated with reactivity of adjacent neoplastic tissues. Staining differences between nonneoplastic and neoplastic mammary tissues were related to the intensity and cytologic distribution of the labeling. Heterogeneous reactivity of morphologically similar cells was documented in nonneoplastic and neoplastic breast epithelial cells as well as in nodal and extranodal breast carcinoma metastases. Immunohistologically detectable antigen was not correlated with prognostic factors such as histologic grade or nodal status. A retrospective study of T1NO cases failed to substantiate any prognostic value for the reactivity of primary breast tumors with this monoclonal antibody.     

Are keratin proteins a better tumor marker than epithelial membrane antigen? A comparative immunohistochemical study of various paraffin-embedded neoplasms using monoclonal and polyclonal antibodies

Epithelial membrane antigen and keratin proteins represent markers of epithelial differentiation that may be detected in routine formalin-fixed, paraffin-embedded tissues. Eighty-seven neoplasms, including 48 adenocarcinomas of various types, squamous and transitional cell carcinomas, small-cell anaplastic carcinomas, carcinoid tumors, mesotheliomas, hepatomas, melanomas (metastatic), adrenal cortical carcinomas, germ cell tumors, and extramammary Paget's disease, were assessed to determine the relative effectiveness of these antigens as tumor markers. Immunoperoxidase studies were performed using monoclonal antibodies to epithelial membrane antigen and monoclonal (combined AE1 and AE3) and polyclonal (bovine muzzle keratins) antibodies to keratin proteins. In more than half the cases (50/87%), both markers yielded comparable results. However, in 29 cases (33%), keratin proteins were clearly superior to epithelial membrane antigen as a tumor cell marker. Particular discrepancies were apparent for some gastrointestinal adenocarcinomas, squamous cell carcinomas, hepatomas (hepatocellular type), spindle cell components of mesotheliomas, and carcinoid tumors. Epithelial membrane antigen represented a better marker in eight cases (9%), mainly for small-cell anaplastic carcinomas and some renal cell and pulmonary adenocarcinomas. Adrenal cortical carcinomas, melanomas, and seminomas were nonimmunoreactive for both antigens. Epithelial membrane antigen and keratin proteins represent useful complementary markers in diagnostic surgical pathology.     

DNA polymerase alpha. An immunohistochemical marker for proliferating cells in normal and neoplastic human tissues.

DNA polymerase alpha is a DNA replicating enzyme expressed in all proliferating cells. This nuclear antigen in paraformaldehyde-postfixed frozen sections of normal, benign, and malignant tissues was identified by the peroxidase-antiperoxidase technique with the use of a mouse monoclonal antibody against bovine/human DNA polymerase alpha. The nuclei of normal proliferating cells were positive. Malignant tumors (n = 95) showed a higher proportion of positive nuclei than did low-grade malignant tumors (n = 7) or benign lesions (n = 67). The number of positive nuclei in squamous cell carcinomas (n = 19) was higher than in adenocarcinomas (n = 45). Eight (18%) adenocarcinomas and all five renal cell carcinomas had less than 10% positive cells, whereas in benign tissues, such as pituitary adenomas, a thymoma, reactive lymphoid lesions, and some benign mammary nodules, more than 10% of nuclei were labeled. In addition, foci of proliferating cells were clearly recognized. DNA polymerase alpha is, therefore, an excellent marker of proliferative activity that provides an approach to analyzing tumor cell heterogeneity not only in fully developed neoplasms, but also in their precursor lesions.